3D perception & embodied AI

Yi Zhang

I am a PhD student in the Department of Electrical and Electronic Engineering at The Hong Kong Polytechnic University, supervised by Prof. Lap-Pui Chau (IEEE Fellow). I work closely with Prof. Wang Yi. My research interests include image-based 3D perception and embodied artificial intelligence.

I received my B.Sc. in Computer Science from the Hong Kong University of Science and Technology in 2020, and my M.Sc. in Data and Artificial Intelligence from Institut Polytechnique de Paris in 2023.

Research

ORCID record

I study how images can support 3D scene understanding, with a focus on object detection, visual grounding, and Gaussian representations.

Under review

Select, Compete, Refine: Training-Free 3D Visual Grounding from RGB-D Views

2026

SeqAlign3DVG: A Sequence-Aligned Benchmark and Voxel Reasoning Framework for 3D Visual Grounding

Yi Zhang, Yi Wang, Yueting Wu, Kaiyue Yang, Yuejiao Su, Lap-Pui Chau

ACM Multimedia (ACM MM), 2026

An observation-aligned grounding benchmark and a voxel reasoning framework for single images and ordered RGB sequences.

arXiv
Code

GVSynergy-Det: Synergistic Gaussian-Voxel Representations for Multi-View 3D Object Detection

Yi Zhang, Yi Wang, Lei Yao, Lap-Pui Chau

IEEE Transactions on Circuits and Systems for Video Technology (TCSVT), 2026

Combining Gaussian surface features and voxel context for multi-view 3D detection without depth or dense 3D geometry supervision.

MASS: Mesh-inellipse Aligned Deformable Surfel Splatting for Hand Reconstruction and Rendering from Egocentric Monocular Video

Haoyu Zhu, Yi Zhang, Lei Yao, Lap-Pui Chau, Yi Wang

Computational Visual Media (CVM), 2026 · Journal Track

Mesh-aligned deformable Gaussian surfels for detailed hand reconstruction and rendering from egocentric monocular video.

arXiv

2025

GaussianCross: Cross-modal Self-supervised 3D Representation Learning via Gaussian Splatting

Lei Yao, Yi Wang, Yi Zhang, Moyun Liu, Lap-Pui Chau

ACM Multimedia (ACM MM), 2025

Learning 3D representations through Gaussian splatting and joint distillation of appearance, geometry, and semantics.

3DGeoDet: General-Purpose Geometry-Aware Image-Based 3D Object Detection

Yi Zhang, Yi Wang, Yawen Cui, Lap-Pui Chau

IEEE Transactions on Multimedia (TMM), 2025

Explicit voxel occupancy and implicit surface geometry improve 3D detection from single-view and multi-view images.

A Survey of Embodied Learning for Object-Centric Robotic Manipulation

Ying Zheng, Lei Yao, Yuejiao Su, Yi Zhang, Yi Wang, Sicheng Zhao, Yiyi Zhang, Lap-Pui Chau

Machine Intelligence Research, 2025

A review of embodied perception, policy learning, and task-oriented learning for object-centric robotic manipulation.

Yi Zhang

Research

Select, Compete, Refine: Training-Free 3D Visual Grounding from RGB-D Views

2026

SeqAlign3DVG: A Sequence-Aligned Benchmark and Voxel Reasoning Framework for 3D Visual Grounding

GVSynergy-Det: Synergistic Gaussian-Voxel Representations for Multi-View 3D Object Detection

MASS: Mesh-inellipse Aligned Deformable Surfel Splatting for Hand Reconstruction and Rendering from Egocentric Monocular Video

2025

GaussianCross: Cross-modal Self-supervised 3D Representation Learning via Gaussian Splatting

3DGeoDet: General-Purpose Geometry-Aware Image-Based 3D Object Detection

A Survey of Embodied Learning for Object-Centric Robotic Manipulation

Service

Teaching assistant

Reviewer

Conferences

Journals